Metraipolis › archive › dc_adb3ca5179cf

DyadMem: A Long-Term Memory Benchmark of How Agents Work with Users — arXiv:2610.03020

Source
arXiv
Original
https://arxiv.org/abs/2610.03020
License (as archived)
CC0-1.0
Archived
2026-10-05T22:49:27.236Z
SHA-256 of the archived text
a4dff3e4b3366f73ae94c3247035ab15ad591115488391c4f53e2732e8316130
Stable URL of this page
https://metraipolis.com/documento/dc_adb3ca5179cf

How to cite

DyadMem: A Long-Term Memory Benchmark of How Agents Work with Users — arXiv:2610.03020 — arXiv, https://arxiv.org/abs/2610.03020 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_adb3ca5179cf (Metraipolis archive dc_adb3ca5179cf, sha256 a4dff3e4b3366f73ae94c3247035ab15ad591115488391c4f53e2732e8316130)
BibTeX
@misc{dcadb3ca5179cf,
  title = {DyadMem: A Long-Term Memory Benchmark of How Agents Work with Users — arXiv:2610.03020},
  howpublished = {arXiv},
  url = {https://arxiv.org/abs/2610.03020},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_adb3ca5179cf; sha256 a4dff3e4b3366f73ae94c3247035ab15ad591115488391c4f53e2732e8316130},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv, https://arxiv.org/abs/2610.03020. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Long-term agents must remember not only what is true about a user, but also how a particular agent should work with that user as their shared history evolves. Existing benchmarks primarily supervise user facts and preferences or experience reusable across users, leaving this relationship-specific agent memory implicit. Additionally, most prior works measure the model solely with final-answer QA over long interaction histories, making the assessment still incomplete and unreliable. To this end, we introduce DyadMem with the proposed new definition User-conditioned Relational Agent Memory (URAM). DyadMem jointly annotates user-side memory and URAM along the same multi-session trajectories, resulting in 6 memory categories. To summarize, it includes 3,065 episodes, 50,961 sessions, and 61,210 QA instances, with extensive session-level Capture and Update gold annotations, query-level Recall support, and two QA settings: Gold-Memory and Full-Pipeline. Across 16 open-weight and 4 proprietary models, Gold-Memory QA is consistently strong, yet Full-Pipeline QA drops sharply. Such a gap explicitly supports our fine-grained evaluation design. Additionally, several quantitative results further reveal low Capture recall, incomplete Recall, and unsafe-deletion issues arising from even the frontier LLMs. We further conduct a rigorous experiment to validate the effectiveness of our URAM and observe the positive effects for all 20 models. In summary, DyadMem is a dual-domain, full-pipeline memory benchmark with extensive annotation efforts for advancing the domain's development.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2610.03020 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-05T22:45:55.666Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).